Brian.......I understand the point you are making. There are a number of variables that impact amplifier and speaker performance, as well as volume level and dynamics. The first thing that comes to mind for me is the quality and robustness of a tube amplifier's power supply, and also the particular type of power tubes in use. If an amplifier's power supply is designed to deliver 3dB of momentary headroom over its rated power, and there aren't many amplifiers that do, the power supply would be required to momentarily deliver twice its design output capability for a very brief moment, while still maintaining its ability to deliver the continuous power rating. This is a critical point because the more robust the power supply and reserve capacitance built into the power supply is, the more likely momentary high current demands can be delivered without shortfall. What we hear is clean startling dynamics, or weak compressed dynamics. Any amplifier, and that includes high powered solid state amplifiers, can be driven to their limits, and when this takes place many irregularities show their impact on the sound. As for tube amplifiers, the type and number of power tubes that generate an amplifier's output differ with circuit designs. Some tubes cannot produce high current demands no matter what the power supply is capable of doing, while other power tubes can readily rise to momentary high current demands in stride. When driving speakers with a relatively benign impedance curve, high current reserves are not as important as when driving speakers with huge impedance swings, especially in the lowest two octaves where amplifiers deliver their highest output current. It really is a complex issue that fluctuates from one speaker/amplifier package to the next. The variables are almost infinite.
Then there is the issue of volume level and what anyone considers their normal listening level. My comments above are based on my personal listening habits, which by the way have toned themselves down quite a bit as I grow older. Lots of folks like to rock the room, and this can require additional power or not depending on the efficiency of the speakers. I once owned a pair of old Alec Voice of the Theater speakers. These speakers were 16 ohm and rated at 102dB/1W/1M. My 30 watt tube mono block McIntosh MC30 amplifiers never broke a sweat even at window rattling, furniture vibrating sound pressure levels. You commented that your Dyn's at 6 ohms had an efficiency in the mid 80 decibels (assuming 85dB). Where 1 watt would produce 102dB of sound pressure with the old Altecs, the Dyn's would require approximately 64 watts to produce that same 102dB of sound pressure. Depending on how the Dyn's impedance curve looked at the lowest frequency range the CJ amp may just not have been the right match from the beginning. If the music had 6dB of dynamic range above average music level then the most power you could safely use without clipping the amplifier would be in the range of 16 watts output in order to have enough reserve power for the 6dB peaks.
I reiterate, the variables of amplifier and speaker combinations are huge, as are the personal listening habits of all of us. What works for one may turn out to be totally wrong for the next person. I have experimented with many McIntosh amplifiers for over 50 years, both tube and solid state. One of my primary reasons for continuing to own McIntosh amplifiers is their philosophy of over-designing there amps. The MC75 is a perfect example. It uses the same power supply for its mono chassis as the MC275 stereo amp uses for both channels. I suspect this is one of the reasons so many people have said they prefer the MC75 over the MC275, and some have even said the prefer the MC75 over the MC2301 tube power amps. I can only speak from my personal experience, but I have never had issues with dynamic range. I did notice that the MC275's I owned, and I owned three of them, did give up a certain tautness to the lowest octave range that has never been an issue with my MC2301's. From what I have read the McIntosh MC75's more robust power supply on a mono chassis solves this issue completely. We shall find out soon enough.
As for the accuracy of power amplifier meters I can only repeat what McIntosh says about their meters. The manuals on amplifiers with meters states: Power Output Meter The McIntosh MC601 has a large Output Watt Meter that responds 95% full scale to a single cycle tone burst at 2kHz. Voltage and current outputs are electronically measured, multiplied and fed to a special circuit that accelerates the pointer movement in the upward direction. When the pointer reaches its peak it pauses only long enough for the human eye to perceive its position, then drops. It is almost 10 times faster than a professional VU meter. A front panel switch is provided to change the meter to the Watts Hold Mode of operation. This allows fast upward movement of the pointer but greatly increases Hold Time at the peak of its travel. The highest power output of the source material is thus recorded.